ABHISHEKKUMAR.AI

Available for new opportunities

Senior GenAI Engineer.

I am a Senior Generative AI Engineer combining 10+ years of software and data engineering experience with 3+ years of deep specialization in LLM orchestration, multi-agent systems, and production-grade RAG architecture.

10+Years Software/Data Exp
3+Years Specializing in GenAI
5+GenAI Applications Deployed to Production

Implemented Projects

Compound Library Design
AI / Drug Discovery

Compound Library Agent

Production multi-agent LLM platform for drug discovery, implementing memory-augmented reasoning layers and agent coordination using LangGraph.

LangGraphFastAPIRedisAWS EKS
Enterprise RAG Platform
AI / Enterprise

Enterprise RAG Platform

Scale-ready Generative AI RAG platform supporting multiple business units with user-aware document access control and hybrid retrieval.

LangChainAWS BedrockPineconeEKS
ChatEDA Platform
AI / R&D Analytics

ChatEDA Analytics

Production LLM-powered analytics platform for pharma R&D, translating natural language queries into verified code execution and plots.

PythonFastAPIRDKitPandas

Experience

I have been working across GenAI, data systems, and cloud architecture. Here is a timeline of my journey.

Apr 2025 - Present

Senior GenAI Engineer · Syngene International | Bristol Myers Squibb

Architecting and deploying production-grade multi-agent LLM platforms and scalable inference services for drug discovery and pharma R&D.

  • Deployed production multi-agent platform for drug discovery using LangGraph to support complex scientific reasoning and memory-augmented layers.
  • Led development of ChatEDA, an LLM-powered analytics platform for pharma R&D data exploration with RDKit and controlled tool invocation.
  • Designed supervisor-orchestrated agent workflows for High-Throughput Screening (HTS) of 3M compounds, narrowing candidates to 3,000 hits.
  • Migrated LLM inference from SageMaker endpoints to containerized autoscaling GPU EKS clusters, improving cost-efficiency by 60–70%.
LangGraphFastAPIRedisAWS EKSLangSmith
Nov 2023 - Mar 2025

LLM Engineer · Eli Lilly and Company

Architecting enterprise-grade Generative AI RAG platforms with full LLM orchestration, hybrid retrieval, and governance.

  • Defined standards for embedding model selection, retrieval strategies, and deployment patterns using LangChain, Bedrock, and EKS.
  • Designed intelligent retrieval and access control layers with user-aware document access and role-based retrieval.
  • Built ingestion pipelines connecting multi-format documents and APIs into Pinecone and Elasticsearch, reducing vector infrastructure costs by ~90%.
  • Established automated RAG evaluation and monitoring frameworks (LangSmith/TruLens), reducing manual validation effort by ~40%.
LangChainAWS BedrockPineconeElasticsearchTruLens
Sep 2020 - Jun 2023

Data Engineer · Biofourmis

Building real-time data streaming platforms and scalable ETL pipelines for IoT and medical sensor health monitoring.

  • Led development of real-time PySpark data streaming pipelines on EKS for low-latency processing.
  • Developed near real-time ingestion pipelines for medical and IoT sensor data using Redshift and Lambda.
  • Designed robust ETL workflows across heterogeneous database systems and managed Airflow orchestration reliability.
PySparkAWS RedshiftLambdaAirflowPostgreSQL
Aug 2019 - Aug 2020

Software Engineer · Maplelabs

Building data engineering pipelines for stock market analytics and migrating monolithic web applications.

  • Built Python ingestion pipelines for stock market data and trading backtesting strategies for a US-based startup (DCM).
  • Modernized monolithic platform architecture by refactoring into microservices using Python and AWS.
  • Developed Snappy Flow "Poller" Django backend services with Next.js/React and Redux frontend.
PythonDjangoReactNext.jsAWS
Oct 2015 - Aug 2019

Application Development Analyst · Accenture

Deploying machine learning models and data governance pipelines for enterprise analytics.

  • Deployed machine learning models as REST APIs using Python, R, Django, and Docker under an MLOps framework.
  • Engineered data governance pipelines for the IDAG (GDPR-as-a-Service) platform using Django and PySpark.
  • Designed Flask-based ingestion services for NBN Australia with structured storage in PostgreSQL.
PythonDjangoDockerPySparkPostgreSQL

Education

June 2011 - June 2015

B.E. in Information Science & Engineering · Dayananda Sagar College of Engineering

VTU University · Bengaluru, India